Under the assumption of isotropic turbulence, u' = v' = w'

Since the turbulent kinetic energy: k = 0.5 (u'^2 + v'^2 + w'^2), then under this condition if you know k then u' = sqrt(2k/3).

Hello,

If you go through the modeling of turbulence in a RSM model (6 equations for Rij + 1 equation for dissipation) , the instantaneous fluctuations (u', v', w') are inaccessible. Although you can get the magnitudes by taking square root of R11, R22 and R33. 

Hope it helps !

@Vilas Thanks for the answer.

You can use the DPM (multiphase model), inject flow tracers into your domain and extract the fluctuating part of the velocity by writing your flow tracers data.

Under the assumption of isotropic turbulence, u' = v' = w'

Since the turbulent kinetic energy: k = 0.5 (u'^2 + v'^2 + w'^2), then under this condition if you know k then u' = sqrt(2k/3).

If you check the menu of printable variables you can print velocity derivatives and turbulent viscosity values for all of the domain. For RANS models (or RAS) then, you can compute  estimatively fluctuations from Boussinesq assumption and get some idea of your CFD modeling comparing these against experimental data, see any book  of turbulence or http://www.cfd-online.com/Wiki/Boussinesq_eddy_viscosity_assumption

As the elliptic boundary conditions are required for RSM transport equation ,the fluctuated velocity components at solid wall can be calculated from:

u'1=sqrt(1.1k) & u'2=sqrt(0.25k) & u'3=sqrt(0.66k)

Dear Vilas,

I have just downloaded my paper "Statistical model of turbulent nature flow" at RG site. There are relationships for turbulent  velocity components u', v', w', which you  may use in your calculations.

Regards,

Elena Dolgopolova